关于「盒计数法的度量学:维数到底量了什么」,《书法运笔平滑度与分形维数》给出的定量结论是什么?

结论:第五章校验盒计数法,并说明笔宽从 0.5 到 5.0 像素就把维数从 1.0635 推到 1.5777,远大于两组之间 0.015 的差。

数据口径:书中轨迹、纸墨与评分参数均为示意;全部计算由书中程序生成;本书是非个人化的分析,不构成经营、设计或医疗建议。 第二章证明盒计数维数看不见运笔:同一条弧线以三种时序写出,维数都是 1.2251,平滑度却从 -7.653 落到 -11.140;示意的两组笔迹中,维数的效应量只有 -0.36(95 % 区间 -0.98 到 +0.27),平滑度的效应量是 3.74;随机游走的维数 1.704 已接近滴画研究报告的 1.72。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.015, 0.5, 1.0635, 1.5777, 5.0

factor variedvalueD 2-32
brush radius px0.51.0635
brush radius px1.51.2738
brush radius px3.01.4496
brush radius px5.01.5777
flying-white share0.01.2738
flying-white share0.31.2529
flying-white share0.61.1985
flying-white share0.81.0591
strokes in character21.1188
strokes in character41.1965
strokes in character81.2738
strokes in character161.4149
for scale: group gap in p210.015
查看完整验算程序
# Chapter 5. p52: what the dimension of a written character tracks
# Model: a synthetic "character" of k strokes, each a gently curved
# quadratic Bezier stroke placed at random on a 256 px canvas, inked
# with a disc of radius rho; flying white removes a share f of the
# ink pixels at random. D is the box-counting slope over boxes of 2 to
# 32 px, as in p21 and p22. We vary one factor at a time from a base
# of k = 8, rho = 1.5, f = 0. The canvases are synthetic; the gap of
# 0.015 between group means in p21 is printed for scale. Seed fixed.
import math
import random

BOXES = (2, 4, 8, 16, 32)
P21_GAP = 0.015


def slope(pix):
    xs = [math.log(1.0 / s) for s in BOXES]
    ys = [math.log(len({(x // s, y // s) for x, y in pix}))
          for s in BOXES]
    mx, my = math.fsum(xs) / len(xs), math.fsum(ys) / len(ys)
    sxy = math.fsum((a - mx) * (b - my) for a, b in zip(xs, ys))
    return sxy / math.fsum((a - mx) ** 2 for a in xs)


def strokes(k):
    rng = random.Random(20261113)
    out = []
    for _ in range(k):
        a = (rng.uniform(40, 216), rng.uniform(40, 216))
        b = (rng.uniform(40, 216), rng.uniform(40, 216))
        m = ((a[0] + b[0]) / 2 + rng.uniform(-25, 25),
             (a[1] + b[1]) / 2 + rng.uniform(-25, 25))
        out.append((a, m, b))
    return out


def render(strk, rho, f):
    rng = random.Random(20261114)
    r = int(math.ceil(rho))
    disc = [(dx, dy) for dx in range(-r, r + 1) for dy in range(-r, r + 1)
            if dx * dx + dy * dy <= rho * rho]
    pix = set()
    for a, m, b in strk:
        for i in range(801):
            t = i / 800
            x = (1 - t) ** 2 * a[0] + 2 * t * (1 - t) * m[0] + t * t * b[0]
            y = (1 - t) ** 2 * a[1] + 2 * t * (1 - t) * m[1] + t * t * b[1]
            for dx, dy in disc:
                pix.add((int(round(x)) + dx, int(round(y)) + dy))
    return {p for p in sorted(pix) if rng.random() >= f}


if __name__ == "__main__":
    base = strokes(8)
    print("factor varied          value     D 2-32")
    for rho in (0.5, 1.5, 3.0, 5.0):
        print("brush radius px        {:>5.1f}    {:.4f}".format(
            rho, slope(render(base, rho, 0.0))))
    for f in (0.0, 0.3, 0.6, 0.8):
        print("flying-white share     {:>5.1f}    {:.4f}".format(
            f, slope(render(base, 1.5, f))))
    for k in (2, 4, 8, 16):
        print("strokes in character   {:>5d}    {:.4f}".format(
            k, slope(render(strokes(k), 1.5, 0.0))))
    print("for scale: group gap in p21  {:.3f}".format(P21_GAP))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《书法运笔最小加加速度轨迹、飞白墨量毛细流与笔势美感分形维数母典》第 5 章「盒计数法的度量学:维数到底量了什么」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-229 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026229,
  author    = {Fan, Yafu},
  title     = {Minimum-Jerk Brush Trajectories, Capillary Ink Flow in Flying-White Strokes and the Fractal Dimension of Stroke Beauty},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论